longfellow bridge historic rehabilitation · without challenges. rivets the longfellow bridge...

10
Page 1 of 10 LONGFELLOW BRIDGE HISTORIC REHABILITATION Abstract The Longfellow Bridge is a well-known Boston and Cambridge landmark. The structure, which carries both roadway traffic and rail, has recently been rehabilitated through a $305 million Design-Build Contract. Built in 1907, the bridge has become an emblem for the Commonwealth. The Rehabilitation of the Longfellow Bridge set out not only to re-establish the functionality of a critical piece of Massachusetts infrastructure, but also to preserve the original bridge aesthetic appearance. The paper below provides a brief overview of the project and a description of the steel detailing and fit up challenges which involved the installation of over 200,000 new pieces of steel. Introduction The historic Longfellow Bridge spans the Charles River connecting Boston and Cambridge. Stretching more than 1,900 feet, it carries Route 3 and the Massachusetts Bay Transportation Authority’s (MBTA) Red Line subway over the river. Having served motorists, pedestrians, and rail travelers for more than 100 years, the bridge had widespread deterioration of its arches, columns, ornate masonry and unique metal casting features. The rehabilitation project improved the structural integrity of the bridge while restoring its distinctive historic architectural features. MassDOT Highway Division contracted the Design-Build team of the WSC joint venture of J.F. White Contracting Company/Skanska/Consigli Construction Co. (WSC) and STV, as the lead designer, for the $305 million project. STV was a sub-consultant to the WSC joint venture. The project encompassed the complete reconstruction of the bridge’s original 11 arch spans; a 12th span installed in the 1950s; the seismic retrofit of 12 masonry substructures; and the dismantling, repair and reconstruction of the four signature “salt and pepper” granite towers flanking the main span. The name stems from the 58-foot towers’ resemblance to salt and pepper shakers. The project was part of MassDOT’s Accelerated Bridge Program [1] , whose goal is to reduce the state’s backlog of bridges in need of rehabilitation. .

Upload: others

Post on 04-Apr-2020

5 views

Category:

Documents


0 download

TRANSCRIPT

Page 1 of 10

LONGFELLOW BRIDGE HISTORIC REHABILITATION

Abstract

The Longfellow Bridge is a well-known Boston and Cambridge landmark. The structure, which carries both

roadway traffic and rail, has recently been rehabilitated through a $305 million Design-Build Contract. Built

in 1907, the bridge has become an emblem for the Commonwealth. The Rehabilitation of the Longfellow

Bridge set out not only to re-establish the functionality of a critical piece of Massachusetts infrastructure, but

also to preserve the original bridge aesthetic appearance. The paper below provides a brief overview of the

project and a description of the steel detailing and fit up challenges which involved the installation of over

200,000 new pieces of steel.

Introduction

The historic Longfellow Bridge spans the Charles

River connecting Boston and Cambridge.

Stretching more than 1,900 feet, it carries Route 3

and the Massachusetts Bay Transportation

Authority’s (MBTA) Red Line subway over the

river. Having served motorists, pedestrians, and

rail travelers for more than 100 years, the bridge

had widespread deterioration of its arches,

columns, ornate masonry and unique metal casting

features.

The rehabilitation project improved the structural

integrity of the bridge while restoring its

distinctive historic architectural features.

MassDOT Highway Division contracted the

Design-Build team of the WSC joint venture of

J.F. White Contracting

Company/Skanska/Consigli Construction Co.

(WSC) and STV, as the lead designer, for the $305

million project. STV was a sub-consultant to the

WSC joint venture.

The project encompassed the complete

reconstruction of the bridge’s original 11 arch

spans; a 12th span installed in the 1950s; the

seismic retrofit of 12 masonry substructures; and

the dismantling, repair and reconstruction of the

four signature “salt and pepper” granite towers

flanking the main span. The name stems from the

58-foot towers’ resemblance to salt and pepper

shakers.

The project was part of MassDOT’s Accelerated

Bridge Program [1]

, whose goal is to reduce the

state’s backlog of bridges in need of rehabilitation.

.

Page 2 of 10

Figure 1: Longfellow Bridge after completion of rehabilitation in 2018.

Figure 2: Cross-sections of the existing bridge (top) and proposed rehabilitated bridge (bottom) with

four construction phases.

The Bridge

The Longfellow Bridge was a product of the City

Beautiful Movement, a reform philosophy of

North American architecture and urban planning

that flourished during the 1890s and 1900s with

the intent of introducing beautification and

monumental grandeur in cities. Prior to designing

the Longfellow, the original designers of the

bridge, William Jackson, chief engineer, and

Edmund M. Wheelwright, consulting architect,

travelled to several European cities where they

inspected notable bridges, such as the Pont du

Midi in Lyons, France and the bridge over the

Moskva River in Moscow[2]

. Their design for the

Longfellow consisted of 11 steel arch spans

supported on ten granite block masonry piers and

two massive abutments. Two piers have sculptures

that represent the prows of Viking ships. The two

central piers carry the neoclassically inspired

granite towers. The arch spans increase in length

successively from each bank out to the main span

at the center of the Charles River. The end spans

measure 100 feet in length whereas the main span

measures 188 feet.

Historic Review

The Longfellow Bridge is considered

Massachusetts’ most historically significant

Page 3 of 10

bridge, making it subject to federal preservation

standards and reviews. To head off potential

scheduling snags, the project team implemented a

consultation process at the outset of the project to

expedite issues regarding historical aspects of the

bridge rehabilitation with the six federal, state and

local stakeholders involved under Section 106 of

the National Historic Preservation Act (NHPA).

NHPA gives consulting parties the authority to

review all aspects of the design that may affect the

bridge’s historic character. As designs developed,

the design-build team met regularly with

preservation officials to outline constraints,

describe possible options and provide

recommendations. The team used 3-D software to

show the project’s visual impacts to the

preservation officials.

STV ultimately obtained design approval from

MassDOT, the Federal Highway Administration,

Massachusetts Bay Transportation Authority,

Massachusetts Department of Conservation and

Recreation, U.S. Coast Guard, the City of Boston,

the City of Cambridge, the Massachusetts

Department of Environmental Protection and the

Historic Review Board (Section 106).

Structural Analysis

The original MassDOT contract called for the re-

use of the 122 steel arch ribs that support the

original 11 spans of the bridge. However, due to

capacity concerns, the contract allowed for up to

12 of the arches to be replaced. The design team

set a goal of preserving all of the arch ribs.

National Bridge Inspection Standards-trained

inspectors conducted a series of hands-on

inspections and nondestructive testing of the steel

arch ribs. They recorded section loss and any other

signs of structural distress and delivered the data

to the design team.

A full 3-D analysis model was developed for each

span, and several thousand live load cases were

investigated, which combined roadway, pedestrian

and rail traffic. Through this process of using

actual and exact measured arch properties and

extensive modelling, the team demonstrated that

all of the arches had adequate capacity to meet

current code requirements.

Figure 3: 3-D Model used to analyze arches for

various live load cases, including pedestrian,

roadway and train loads.

Construction Phasing

A crucial transportation link between Boston and

Cambridge, the Longfellow Bridge carries 90,000

Red Line train passengers, 28,000 cars and trucks

and well over 1,000 bicyclists and pedestrians

daily between the cities[3]

. Since it is such a key

thoroughfare, it was essential to maintain the Red

Line service and minimize impacts to the traveling

public, local businesses and the boating

community during construction.

When WSC and STV reviewed the proposed

construction staging approach in the contract

documents, the team identified several safety and

feasibility concerns. The design-build team

developed a staging plan that allowed for a

sequencing of work that progressed systematically

from the upstream side of the bridge to the

downstream side, and which simplified the steel

fit-up. More importantly, the construction stage

bounded by active rail traffic on both sides was

eliminated, which increased worker safety and

productivity and reduced the number of activities

requiring weekend closures.

The phasing schemes developed by the team were

instrumental in keeping the Red Line open and

traffic moving. There were four construction

phases, but it took six traffic stages to perform the

four construction phases. The STV team

Page 4 of 10

developed plans to maintain access for all users as

well as emergency vehicles. The MBTA Red Line,

sidewalks, bike lanes and one inbound (toward

Boston) vehicle travel lane were kept open on the

bridge at all times. Extensive detour plans allowed

for traffic movement around the Charles River

Basin between Boston and Kendall Square for

vehicular outbound traffic (towards Cambridge).

During 25 approved Red Line shutdown

weekends, MBTA customers used bus service.

The team worked closely with the Cities of Boston

and Cambridge as well as numerous stakeholder

groups to reduce impacts to bridge users.

Figure 4: Construction Stage 2 with train on

shoo-fly during work under MBTA reservation.

For the first phase of construction, the Boston-

bound (inbound) roadway was closed, inbound

traffic was diverted to the outbound roadway on

the downstream side and outbound traffic was

detoured to the river crossing at the Museum of

Science. During the second phase, both the

inbound side roadway and inbound Red Line track

were closed. Inbound MBTA trains shifted to the

outbound track and outbound trains were displaced

to a temporary “shoo-fly” bypass track placed on

the outbound roadway. Stages 3 and 4 mirrored

Stages 1 and 2 for the outbound side of the bridge.

Through all phases, the bridge was kept open to

Red Line trains, pedestrians, cyclists and inbound

vehicular traffic.

On the bridge’s Boston side, the rail-supported

structure in the MBTA reservation (the area

between the fences with the MBTA’s tracks) on

Spans 2 and 3 could not be accessed, even with the

use of the shoo-fly track. This length of structure

had to be replaced through accelerated

construction during weekend closures of MBTA

Red Line service. STV produced a design scheme

for the accelerated construction that limited the

closures to just six weekends for 500 linear feet of

track-supported structure replacement.

Detailing and Fit Up Challenges

The historic nature of the bridge drove many of

the design challenges. In addition to the project

being compliant with modern engineering

practices and codes, the historic restoration and

preservation of certain elements was of vital

importance.

Based on the Section 106 “Conditional No

Adverse Effect” finding for the project, the

structure was broken into three categories, namely:

Critical Elements to be Restored

Elements to be Sensitively Rehabilitated

Elements of Little/No Historic Value.

Page 5 of 10

Figure 5: Arch span including historically

critical railing and fascia and sensitively

rehabilitated arch ribs and spandrel columns.

Critical Elements included existing ornamental

cast iron fascia castings and cover plates, and the

existing ornamental pedestrian railing.

Elements to be Sensitively Rehabilitated were

generally the most visible members on the bridge,

including all the arches, the A-line columns along

all spans, and most of the columns and buckle

plates on Spans 1, 2, and 11, which span over

roadways as opposed to the river.

In order to rehabilitate these members with

historic sensitivity, original design detailing was

considered and replicated as necessary, but not

without challenges.

Rivets

The Longfellow Bridge utilized hot riveting during

the original construction, which was a popular

technique used in bridge construction at the time.

Since then, however, rivets have become

essentially obsolete in the construction industry

due to developments within the bolt industry.

Rivets, although cheaper to manufacture, are more

labor intensive and require more specialized skills

and tools than bolts. And bolts have come a long

way throughout history, from becoming

standardized across the world to gaining strength

and consistency between bolts as knowledge and

materials develop.

Around the 1960s and 1970s, bolts took the place

of rivets as the standard fastener type and

construction technique across the bridge industry.

By today, the practice of fabrication of steel

bridges through the use of riveting has essentially

disappeared in the USA.

It proved challenging for the design-build team to

find a steel fabricator with the skilled labor and

tools, or the desire to obtain training and tools, to

work on riveted members for the Longfellow

Bridge rehabilitation. Since the original

construction used rivets, the members designated

to be historically accurate replica members

required riveted construction to match the original

fastener type. These members include all A-line

spandrel columns, sidewalk stringers and sidewalk

beams across all spans. Additionally, any columns

or buckle plates that could not be rehabilitated,

knee braces, and arch rib girder and column

diaphragms on Spans 1, 2 and 11 required

historically accurate replica members fabricated

using rivets.

Members that were designated to have modified

historically accurate replacement members as

opposed to replica members were fabricated using

button-head tension control bolts. These include

the less visible members on Spans 3-10, such as all

column lines except the A-line.

Figure 6: 3-D Model used to show button-head

TC bolts vs hex-head bolts in column

installation.

Due to the dangerous nature of hot riveting, all

riveting was confined to the shop for built-up

members. All field connections between members

were made using bolts. Where possible, button-

head tension control (TC) bolts were used for

connections with the button-head facing towards

the most visible direction of the connection to

maintain a similar aesthetic to the original rivets.

In some instances, such as the inner row of column

connections to the arches, hex-head bolts needed

to be used for installation purposes. Based on the

entering and tightening clearances available with

the column design and the existing rivet holes

being reused for bolts, the torque gun used to

tension the TC bolts did not fit, and hex head bolts

were used instead.

Buckle Plates

Included on the list of historically significant

elements on the bridge to preserve were the buckle

plates. When the bridge was originally constructed

in 1907, steel buckle plates were installed across

the bridge, spanning between stringers and

floorbeams to support the concrete deck and

Page 6 of 10

sidewalk. Two convex “buckles” per panel

provided two-way action to support the

unreinforced concrete slab above. [4]

Figure 7: Existing buckle plates from below

before demolition.

In 1959, during a major rehabilitation of the

bridge, portions of the deck were removed, along

with the supporting buckle plates, and replaced

with sections of reinforced concrete. For the 2018

rehabilitation, the entire deck and sidewalk was

replaced. With a new reinforced concrete, the

buckle plates become obsolete structurally, but

remain important from a historic perspective.

A goal during reconstruction was to salvage and

reuse any of the original buckle plates that were in

good condition, installing the original buckle

plates and any additional replica buckle plates

needed to fill in the three spans over the roadways,

Spans 1, 2 and 11. These spans were so designated

because there is pedestrian access below the spans.

Based on the construction phasing, the Design-

Build team proposed, and the Section 106

Consulting Parties agreed, that it was not feasible

with the schedule to extract, refurbish and reinstall

any of the original buckle plates on the upstream

roadway of Span 11. Replica buckle plates were

fabricated and installed during the early phase

work, with the intent to use all salvaged plates on

Spans 1, 2 and the downstream portion of Span 11.

Upon the continuation of demolition however, all

the existing buckle plates were found to have

advanced deterioration, and none of the original

buckle plates were deemed to be in good condition

or able to be refurbished and reinstalled.

In the end, the roadway decks and sidewalks on

Spans 1, 2 and 11 utilized replica buckle plates as

stay-in-place forms. Spans 3-10 over the river did

not have the same requirements, so instead of

using replica buckle plates as stay-in-place forms,

removable formwork was used to cast both

roadway decks and sidewalks.

Columns

The Longfellow Bridge has over 2,600 spandrel

columns evenly spaced along the arches every 7’-

3” to hold up the deck framing. Each built-up

column consists of two C10 channels, inner web

plates at the top and bottom and lattice bars on all

columns greater than approximately 4 feet tall.

The columns make up an important piece of the

aesthetic of the bridge and are considered

historically important. Many of the columns are

categorized as Elements to be Sensitively

Rehabilitated.

Some of the columns were always intended to be

replaced. Some had considerable deterioration,

especially the D-line columns under the joint

between roadway deck and MBTA reservation.

Others had less deterioration but were in areas that

did not require preservation or sensitive

rehabilitation. These columns were replaced with

new steel.

On Spans 1, 2 and 11, which kept the original

framing plan, all roadway columns were deemed

elements to be preserved. All columns that

required replacement within these spans were

designated to be exact replicas, and consequently

required riveted construction.

On Spans 3-10, all columns other than the A-line

were not held to the same preservation

requirements as Spans 1, 2 and 11. The new

framing on Spans 3-10 was modified slightly from

the original framing, and as such, the new columns

along the D-line on these spans are modified

Page 7 of 10

replica columns, maintaining most of the detailing

from the original columns, but modifying the tops

of the columns to work with the new continuous

floorbeams above. Since these columns are

towards the middle of the bridge and located on

the spans that are less visible from land, modified

replication was allowed instead of historically

accurate replication, and the requirements for

construction method were relaxed on these

columns. Instead of riveted construction, the

interior columns on Spans 3-10 were built up

using button-head tension control bolts.

The columns that received the designation from

the Section 106 Consulting Parties as Elements to

be Sensitively Rehabilitated include the A-line

columns across all spans and all the other roadway

columns on Spans 1, 2 and 11. Aside from any

columns that had severe deterioration, which

needed historically accurate replication, all of

these columns were intended to be refurbished and

reused.

During the first phase of demolition, the A-line

columns were carefully removed and brought off-

site for repair and cleaning. The A-line columns

were generally in good condition with little to no

deterioration and needed only minor repairs. Once

cleaned, the columns were hot-dipped galvanized

and painted. However, within weeks of

galvanizing, it became clear that hot-dipped

galvanizing the steelwork was proving to be

problematic. The columns started bleeding rust at

the rivet locations from in between the plies of

steel at the top and bottom web connection plates.

Figure 8: Riveted steel members on Span 1.

After discussions with the Section 106 Consulting

Parties and the owner about the historic

importance of reusing the original materials

compared to replicating the columns with new

steel to ensure a better service life of the members,

the decision was ultimately made to throw away

all the existing steel and replicate all the columns

using new steel. Each piece of the new column

was hot-dipped galvanized individually before

being used in the built-up member. For the

historically important columns along the A-lines

and on Spans 1, 2 and 11, the columns were

fabricated using riveted construction. The field

connections to the arches below and to the framing

above the columns were made using button-head

TC bolts or hex-head bolts.

Figure 9: Rust bleed on existing column after

hot-dipped galvanization of built-up member.

Steel Fit Up

Aside from the challenges of demonstrating

adequate capacity to allow for the reuse of all of

the arches described previously, a major challenge

of reusing the existing arches, especially through

multiple construction stages, was fit up between

the new and existing steel.

During the rehabilitation of the Longfellow

Bridge, the existing arches remained in place, and

the new columns and deck framing were installed

from the bottom up, attaching new steel piece by

piece to the existing arches. Although the

tolerances of the members of the new framing can

be controlled to great accuracy with modern

fabrication techniques, the tolerance of the field fit

up of those new members to the existing arches

was not as accurate. The supporting arches were

not always at the theoretical design location from

the original construction.

By design, the arches are evenly spaced at 10’-3”

on center beneath the roadways and 5’-0” on

center beneath the MBTA reservation, and the

Page 8 of 10

columns are evenly spaced at 7’-3” on center

along the centerline of the arches.

In reality, the arches are not perfectly straight and

evenly spaced across the spans. Some of the

arches have overall sweeping curves horizontally,

with the entire arch bowing into or away from the

centerline of the bridge. Some arches were slightly

tilted or had slightly skewed webs or top flanges.

As rivets and columns were removed from the

arch flanges, it became clear that the rivet holes at

the column base connection plates did not fully

align with the holes in the existing top arch

flanges. Because rivets are installed in a molten

state, the holes they fit into did not need to fully

align since the molten metal would deform into the

space available. Additionally, the holes did not

always line with the center of the existing rivet

heads.

Figure 10: Gantry used during construction to

place steel roadway framing into best-fit

location based on existing arch misalignments.

All these misalignments, as well as the steel arches

moving up and down with temperature changes,

created a large challenge to obtain accurate and

precise field measurements and to achieve proper

steel fit up during installation.

After interaction with MassDOT, it was agreed

that the primary tool used to build tolerance into

the erection of the steel framework was slotted

holes in slip-critical connections. A connection

could utilize slotted holes as long as at each faying

surface, the slot abutted a standard hole.

The existing rivet holes in the arches were reused

for bolts. Holes were reamed as necessary to be

able to fit a bolt through, and this reamed hole was

considered a slot, even if the dimensions were

outside the standard definitions of slots. This

means that the base plates on top of the arches at

each column location needed standard holes. After

the rivets were removed, field measurements

confirmed the locations of the holes, and new base

plates were drilled to match the field conditions.

Even though the arch ribs were slotted, an

additional slot was used in the same connection at

the column base angles. With the base plates

between the two slotted components, each faying

surface only had one slot. The column base angles

were slotted to give adjustment to the columns in

the longitudinal direction of the bridge. The

columns were installed as close to the theoretical

7’-3” spacing as possible.

Figure 11: 3-D Model used to show slotting in

column base angles for connection to arch rib

to add longitudinal adjustability.

Adjustments in the transverse direction were

achieved at the deck framing level. Slots were

added to floorbeams at each of the column

connections to account for any translation

necessary to correct the misalignments of the

arches and columns below. In a few instances,

floorbeams were lengthened or shortened in

addition to slotting the connections.

Page 9 of 10

Figure 12: The final configuration of the bridge is used by pedestrians, bicycles, automobiles and trains.

Vertical adjustment in the bridge was also a

challenge based on the existing arches. Survey

data was collected on the vertical location of the

arches at the pins and crown as well as numerous

intermediate points along the A-line arches, and a

best-fit parabola was found for each arch. After

adjusting for dead load and temperature effects,

column heights were calculated to place the deck

framing at the proper height for the design profile.

Theoretical modeling of temperature effects on a

large arch structure depends on slightly unrealistic

assumptions, such as the entire arch changing

temperature at the same rate across the entire arch

and each arch being installed as a perfect parabola,

so some imprecision is inherent to modeling. To

account for this, vertical adjustments needed to be

included in the design. Nominal half inch shim

packs were included in the column design height,

and once the columns were installed in the field,

the shims could be removed, or additional shims

could be added to reach the desired height to fit up

the framing. Additional height adjustments were

accounted for with the deck haunches.

A Code Compliant Bridge

The bridge is now AASHTO-compliant [5]

and

Americans with Disabilities Act (ADA)-compliant [6]

. In its final configuration, one outbound travel

lane was eliminated, and bicycle lanes were added

on both the inbound and outbound sides. The

bridge has two inbound travel lanes (the same as

before), one outbound lane, two rebuilt MBTA

Red Line tracks, two bicycle lanes and two

widened sidewalks.

Acknowledgements

The authors would like to thank all participants

involved in the Longfellow Bridge Historic

Rehabilitation, especially the following

organizations.

Owner: Massachusetts Department of

Transportation (MassDOT) and Massachusetts

Bay Transportation Authority (MBTA)

Contractor: J.F. White Contracting Company/

Skanska/Consigli Construction Co. (WSC)

Engineer of Record: STV Incorporated

Page 10 of 10

References

1. MassDOT Accelerated Bridge Program (ABP). Commonwealth of Massachusetts 2018.

https://www.mass.gov/accelerated-bridge-program-abp.

2. Architectural Iron & Steel in the 21st Century: Design and Preservation of Contemporary & Historic

Architecture – Conference, April 4, 2016, MIT, Cambridge, MA. Longfellow Bridge Preservation

Presentation – Presenters: Stacey Donahue, Robert Collari, Wendall Kalsow, Mark Ennis.

3. Commonwealth of Massachusetts Department of Transportation, Technical Provisions, Bridge

Rehabilitation Br. No. B-16-009=C-01-002 (Longfellow Bridge), 2012.

4. City of Boston Printing Department. Report of the Cambridge Bridge Commission and the Report of the

Chief Engineer Upon the Construction of Cambridge Bridge, 1909.

5. American Association of State Highway and Transportation Officials, Standard Specifications for

Highway Bridges, 17th Edition, 2002.

6. Americans with Disabilities Act, Standards for Accessible Design, 2010.